Joint phoneme alignment and text-informed speech separation on highly corrupted speech - Equipe Signal, Statistique et Apprentissage Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Joint phoneme alignment and text-informed speech separation on highly corrupted speech

Résumé

Speech separation quality can be improved by exploiting textual information. However, this usually requires text-to-speech alignment at phoneme level. Classical alignment methods are made for rather clean speech and do not work as well on corrupted speech. We propose to perform text-informed speech-music separation and phoneme alignment jointly using recurrent neural networks and the attention mechanism. We show that it leads to benefits for both tasks. In experiments, phoneme transcripts are used to improve the perceived quality of separated speech over a non-informed baseline. Moreover, our novel phoneme alignment method based on the attention mechanism achieves state-of-the-art alignment accuracy on clean and on heavily corrupted speech.
Fichier principal
Vignette du fichier
ICASSP_2020_paper_HAL.pdf (508.32 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02457075 , version 1 (06-02-2020)

Identifiants

  • HAL Id : hal-02457075 , version 1

Citer

Kilian Schulze-Forster, Clément Doire, Gael Richard, Roland Badeau. Joint phoneme alignment and text-informed speech separation on highly corrupted speech. 45th International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2020), May 2020, Barcelona, Spain. ⟨hal-02457075⟩
315 Consultations
923 Téléchargements

Partager

Gmail Facebook X LinkedIn More